def run_knn(name, x_train, x_test, y_train, y_test, tuned_parameters):
print("Working on {} data...".format(name))
img_name = "images/{}_knn_learning_curve.png".format(name)
img_title = '{} kNN Learning Curve'.format(name)
report_name = "reports/{}_knn_output.txt".format(name)
sys.stdout = open(report_name, "w")
clf = get_optimized_classifier(estimator=KNeighborsClassifier(),
tuned_parameters=tuned_parameters,
x_train=x_train,
y_train=y_train)
best_min_samples = clf.best_params_['n_neighbors']
optimized_clf = KNeighborsClassifier(n_neighbors=best_min_samples)
plot_learning_curve(
optimized_clf,
title=img_title,
file_name=img_name,
X=x_train,
y=y_train,
)
run_optimized(optimized_clf, x_train, y_train, x_test, y_test)
sys.stdout = sys.__stdout__
print("Finished {} knn!".format(name))
print()
def run_sa_nn(name, x_train, x_test, y_train, y_test):
print ("Working on SA NN...")
report_name = "reports/{}_nn_sa_output.txt".format(name)
sys.stdout = open(report_name, "w")
sa_nn = mlrose.NeuralNetwork(hidden_nodes = [11], algorithm = 'simulated_annealing', max_iters = 1000)
run_optimized(sa_nn, x_train, y_train, x_test, y_test)
sys.stdout = sys.__stdout__
def run_rhc_nn(name, x_train, x_test, y_train, y_test):
print ("Working on RHC NN...")
report_name = "reports/{}_nn_rhc_output.txt".format(name)
sys.stdout = open(report_name, "w")
rhc_nn = mlrose.NeuralNetwork(hidden_nodes = [11], algorithm = 'random_hill_climb', max_iters = 1000)
run_optimized(rhc_nn, x_train, y_train, x_test, y_test)
sys.stdout = sys.__stdout__
def run_nn(name, x_train, x_test, y_train, y_test):
print ("Working on BackProp")
report_name = "reports/{}_nn_output.txt".format(name)
sys.stdout = open(report_name, "w")
optimized_clf = MLPClassifier(
max_iter=1000,
alpha=0.001,
hidden_layer_sizes=11,
random_state=99
)
run_optimized(optimized_clf, x_train, y_train, x_test, y_test)
sys.stdout = sys.__stdout__
def run_nn(name, x_train, x_test, y_train, y_test, tuned_parameters, iter_range):
print ("Working on {} data...".format(name))
img_name = "images/{}_nn_learning_curve.png".format(name)
img_title = '{} Neural Net Learning Curve'.format(name)
iter_title = '{} Neural Net Iteration Learning Curve'.format(name)
iter_name = "iteration_curves/{}_nn.png".format(name)
report_name = "reports/{}_nn_output.txt".format(name)
sys.stdout = open(report_name, "w")
clf = get_optimized_classifier(
estimator=MLPClassifier(),
tuned_parameters=tuned_parameters,
x_train=x_train,
y_train=y_train
)
best_params = clf.best_params_
optimized_clf = MLPClassifier(**best_params)
plot_learning_curve(
optimized_clf,
title=img_title,
file_name=img_name,
X=x_train,
y=y_train,
)
run_optimized(optimized_clf, x_train, y_train, x_test, y_test)
plot_iterations(
optimized_clf,
file_name=iter_name,
title=iter_title,
X=x_test,
y=y_test,
param_name="max_iter",
param_range=iter_range
)
sys.stdout = sys.__stdout__
print ("Finished {} Neural Net!".format(name))
print()
def run_nn(name, x_train, x_test, y_train, y_test):
img_name = "images/{}_nn_learning_curve.png".format(name)
img_title = '{} Neural Net Learning Curve'.format(name)
iter_title = '{} Neural Net Iteration Learning Curve'.format(name)
optimized_clf = MLPClassifier(max_iter=1000,
alpha=0.001,
hidden_layer_sizes=11,
random_state=99)
plot_learning_curve(
optimized_clf,
title=img_title,
file_name=img_name,
X=x_train,
y=y_train,
)
run_optimized(optimized_clf, x_train, y_train, x_test, y_test, name)
def run_ga_nn(name, x_train, x_test, y_train, y_test):
print ("Working on GA NN...")
report_name = "reports/{}_nn_ga_output.txt".format(name)
sys.stdout = open(report_name, "w")
ga_nn = mlrose.NeuralNetwork(
hidden_nodes = [11],
algorithm = 'genetic_alg',
max_iters = 1000,
early_stopping = True,
clip_max = 10,
max_attempts = 10,
mutation_prob=0.15
)
run_optimized(ga_nn, x_train, y_train, x_test, y_test)
sys.stdout = sys.__stdout__
def run_dt(name, x_train, x_test, y_train, y_test):
print("Working on {} data...".format(name))
img_name = "images/{}_decision_tree_learning_curve.png".format(name)
img_title = '{} Decision Tree Learning Curve'.format(name)
report_name = "reports/{}_decision_tree_output.txt".format(name)
sys.stdout = open(report_name, "w")
tuned_parameters = [{
'min_samples_split': list(range(40, 70)),
"min_samples_leaf": list(range(1, 10)),
}]
clf = get_optimized_classifier(
estimator=DecisionTreeClassifier(random_state=99),
tuned_parameters=tuned_parameters,
x_train=x_train,
y_train=y_train)
best_min_samples = clf.best_params_['min_samples_split']
optimized_clf = DecisionTreeClassifier(min_samples_split=best_min_samples,
random_state=99)
plot_learning_curve(
optimized_clf,
title=img_title,
file_name=img_name,
X=x_train,
y=y_train,
)
run_optimized(optimized_clf, x_train, y_train, x_test, y_test)
sys.stdout = sys.__stdout__
print("Finished {} decision tree!".format(name))
print()
def run_boosting(name, x_train, x_test, y_train, y_test, tuned_parameters):
print("Working on {} data...".format(name))
img_name = "images/{}_boosting_learning_curve.png".format(name)
iter_name = "iteration_curves/{}_boosting.png".format(name)
img_title = '{} Boosting Learning Curve'.format(name)
iter_title = '{} Boosting Iterations Learning Curve'.format(name)
report_name = "reports/{}_boosting_output.txt".format(name)
sys.stdout = open(report_name, "w")
clf = get_optimized_classifier(
estimator=GradientBoostingClassifier(random_state=99),
tuned_parameters=tuned_parameters,
x_train=x_train,
y_train=y_train)
best_params = clf.best_params_
optimized_clf = GradientBoostingClassifier(**best_params, random_state=99)
plot_learning_curve(
optimized_clf,
title=img_title,
file_name=img_name,
X=x_train,
y=y_train,
)
run_optimized(optimized_clf, x_train, y_train, x_test, y_test)
plot_boosting_iteration_curve(optimized_clf, iter_name, iter_title, x_test,
y_test)
sys.stdout = sys.__stdout__
print("Finished {} boosting!".format(name))
print()